Download Guideline on the development of new medicinal products for

26 January 2017
Submission of comments on: Guideline on the clinical
investigation of medicinal products for the treatment of
Ulcerative Colitis - CHMP/EWP/18463/2006 Rev.1
Comments from: EFPIA
Name of organisation or individual
Tiia Metiäinen – [email protected]
Please note that these comments and the identity of the sender will be published unless a specific
justified objection is received.
When completed, this form should be sent to the European Medicines Agency electronically, in Word
format (not PDF).
1. General comments
Stakeholder number
General comment (if any)
Outcome (if applicable)
(To be completed by
(To be completed by the Agency) ordered comments by topic
the Agency)
appearance in document
The guidance is comprehensive and incorporates many
of the recommendations made in the review and
comment process on the UC Concept Paper from 2014.
The guidance however omits consideration of Response
Rates in the induction or remission phase of disease
treatment as a primary efficacy endpoint for approval.
This appears to overlook the importance of response to
therapy in the moderate to severe population. It also
does not seem to be aligned with attaining the indication
for “treatment of active ulcerative colitis” as described in
section 5 ‘ Indications/treatment goals’. It is our
position that a pre-defined ‘Response’ criteria can
represent clinically meaningful “treatment of active
ulcerative colitis” and we would recommend CHMP
consider incorporating language into the guidance
relating to ‘Clinical Response’ as not only a secondary
endpoint but a primary endpoint for pivotal registration
trials.
We welcome the availability of these updated guidelines.
However, we have 3 main areas of concern, where the
EMA proposed changes to the guidelines would
dramatically affect the availability of new and potentially
effective medications for Ulcerative Colitis and Crohn’s
disease in the European Union. We view that this is
contrary to the EMA’s mission to “facilitate development
and access to medicines”, leading to “timely patient
access to new medicines”.
1. ‘Maintenance of remission/Prevention of relapse’:
primary endpoint of “maintenance of
corticosteroid-free remission without
surgery throughout at least 12 months”
The Agency’s suggested primary endpoint of
“maintenance of corticosteroid-free remission without
Stakeholder number
General comment (if any)
Outcome (if applicable)
(To be completed by
(To be completed by the Agency) ordered comments by topic
the Agency)
appearance in document
surgery throughout at least 12 months” is a laudable
aspiration, but is not a feasible endpoint for currently
available medications. Mandating this endpoint in the EU
will impose a requirement for very large maintenance
cohorts, with treatment durations of longer than 12
months, making both the size and cost of maintenance
studies unfeasible.
2. Co-primary endpoints
Draft UC Guidance suggests a “Co-Primary” endpoint
approach that considers both symptomatic relief as well
as an effect on the inflammatory process rather than the
more commonly used Composite indices, such as the
Mayo Clinical index. In the FDA’s Ulcerative Colitis:
Clinical Trial Endpoints Guidance for Industry released
Aug 2016, a composite endpoint which includes an
assessment of signs and symptoms and endoscopic
improvement is proposed.
Our views that the diverging definitions of the primary
endpoints could prove to be problematic and a shift to a
more consistent desired primary endpoint definition
between regional guidance would be of benefit to
patients.
3. Design of maintenance trial
Including only remitters in the primary analysis makes
the sample size needed in induction infeasible; the
induction phase is not anticipated to be long enough to
wean patients from steroids, and finally, many patients
that are responders and not remitters at the end of
induction achieve remission by the end of maintenance.
While we are supportive of EMA draft guidance, one topic
that we believe the guidance should discuss in greater
detail is the treatment of patients that have been
previously exposed to other therapies. For sponsors, it
is critical to have clear expectations from the EMA
because the mucosal healing in these hard-to-treat
Stakeholder number
General comment (if any)
Outcome (if applicable)
(To be completed by
(To be completed by the Agency) ordered comments by topic
the Agency)
appearance in document
patients is very likely to be reduced. Because the
definitions of response have been altered, new drugs
that offer incremental benefits to patients who are
without remaining treatment options may no longer be
pursued by sponsors. Historical evidence demonstrates
that improvements in the pharmacological treatment of
patients with ulcerative colitis occurred in small steps,
yet these products were welcomed by patients and
physicians because they represented additional
treatment options even though they may not be
considered “transformative” products.
As the draft guideline states, UC is rare below 10 years
of age. Yet the clinical development is asked to include
patients from 2 years of age. Fully-powered clinical
efficacy studies in the paediatric population 2 – 10 years
of age might not be feasible due to the low patient
numbers. Therefore clarification is needed on what is
expected to be demonstrated in those children.
Comments: It would be helpful to understand if EMA
recommends any specific guidance to be followed when
developing and validating PRO instruments. Examples
are the Good Practice in Outcomes Research from the
ISPOR or other institutions and the U.S. FDA “Guidance
for Industry Patient-Reported Outcome Measures: use in
Medical Product Development to Support Labeling
Claims”.
2. Specific comments on text – Major Comments
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
98-110
102-103
123-124
Comment: Please clarify if the extent of active ulcerative
colitis needs to be assessed at entry into study or if historical
extent of active disease is acceptable
Comment: Histological evaluation might be part of the
differential diagnosis of UC. Our data suggest that more than
80% of subjects with a Mayo endoscopy score of > 2 (current
eligibility criterial applied in biologic trials) present with
histologically active disease. Therefore, we believe that a
Mayo endoscopy score >= 2 is a surrogate for histologically
active disease at screening. We do not believe that a separate
inclusion criterion related to histologic confirmation of UC
activity is necessary.
Clinician feedback has indicated that it is not necessary for
patients to receive 3 months of treatment with thiopurines in
order to know if someone is going to be refractory to those
agents. While achieving the full extent of response to
thiopurines may take 90 days, it is possible to know much
earlier (e.g., within 30 days) if someone is going to respond to
these agents at all.
Proposed Change: Please change “3 months” to “6
weeks.”
125
Comment: A working definition of “Remission” is proposed to
be based on endoscopic mucosal healing, with no or very mild
signs. See proposed change below.
Suggest making definition(s) of remission consistent with
those used in clinical practice and also FDA draft guidance.
Proposed change (if any): The authors should consider
defining a number of different types of remission, including
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
131-132
(1) clinical remission, as suggested above, (2) endoscopic
remission [Mayo endoscopic subscore of 0-1 or absence of
friability and erosions on UCEIS), and (3) histologic remission.
Comment: Why is the indication “treatment of active
ulcerative colitis” and not “treatment of ulcerative colitis”, as it
includes maintenance of remission?
Why only “maintenance of remission” and not “remission or
sustained remission”?
131-141
Proposed change (if any):
In order to obtain an indication for “treatment of active
ulcerative colitis”, efficacy in both “induction of remission” as
well as “maintenance remission or sustained of remission”
should be demonstrated.
Comment: The rationale for requiring separate investigation
of induction and maintenance in order to achieve separate
induction and maintenance claims; and why certain study
designs are acceptable and others are not acceptable, are
unclear in the guidance. This current text also suggests that a
treat-through study design that demonstrates efficacy in both
induction of remission and sustaining remission is not a
suitable design to obtain the label claim of “treatment of
active ulcerative colitis”
While the short-term goal of treatments is to achieve rapid
symptom relief (induction) and the long-term goal is to
maintain control of the disease (maintenance); in clinical
practice there is not a fixed duration induction phase and a
fixed duration maintenance phase. Clinical practice embraces
a more holistic approach, where patients will be treated with
an intervention until it is clearly evident that the intervention
does not result in benefit. With respect to the use of biologic
treatments, the initial assessment of whether there is/ is not
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
sufficient clinical benefit to justify continuing treatment could
take a few months. This timeframe is consistent with the
estimated peak/ steady state of maintenance PKPD effect to
be achieved across different approved MOAs (~12-20 weeks).
If sufficient initial benefit is achieved, patients will continue to
be maintained on that treatment for a longer time, with
ongoing observation to ensure there is sustained benefit.
Enforcement of a strict induction and maintenance study
paradigm (i.e. induction followed by randomization to active
drug maintenance or withdrawal to placebo) without
consideration of the time to achieve optimal PKPD effects will
limit our ability to evaluate the true efficacy potential of a
given MOA, because patients who “are not induced” into
response will not continue into the randomized maintenance
trial. Historically, biologic trials have studied induction
efficacy at time points ranging from 2 weeks to 12 weeks; and
most of these trials have reported that a substantial
proportion of patients may achieve a delayed response to
induction (i.e. the non-randomized population in the
randomized withdrawal maintenance study).
Thus, a treat-through design, which evaluates efficacy from a
population perspective, would provide a much more accurate
assessment of the real efficacy potential of a MOA, both shortterm and long-term.
Additional comments regarding the appropriateness of treatthrough vs. randomized withdrawal maintenance studies are
provided in response to Lines 340-353.
Proposed change (if any): Please address the
appropriateness of a treat through design to demonstrate
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
136-141
138
147-148
152-159
efficacy and at a minimum add an additional sentence at line
132. “A treat-through study design showing efficacy in both
“induction of remission” and “sustained remission” may be
suitable to obtain an indication for “treatment of active
ulcerative colitis”.
In regard to the recommendations favouring separate
induction and maintenance studies, it should be noted that
comparison to standard of care comparators (eg anti-TNF)
using this methodology incurs substantial complexity.
Similarly, when active comparators are used, potentially
nonsensical treatment regimens may be necessary to maintain
study blinding in randomized withdrawal designs. We believe
comparison to SOC in both induction and in maintenance may
be best accomplished using a treat through methodology.
Comment: While a “treat -through” design may be acceptable
the design of the study will have implications for the
indications that can be claimed.
While it is clear that indication for “treatment of active UC”
can be obtained with adequate demonstration of effect in both
induction and maintenance, and that separate demonstration
of efficacy in induction and demonstration of efficacy in
maintenance may lead to separate indication for induction and
maintenance respectively, it is not clear what indication may
be granted when using a treat-through design
Proposed change (if any): Please clarify the type of
indication granted in case of treat-through design.
Also further define “treat-through”
There are drugs in development for UC that do not have a
direct effect on inflammation; it would be helpful if the
guidance would consider other mechanisms that may have a
benefit to patients (e.g., drugs that directly promote mucosal
healing)
Comment: The revised guideline encourages use of a PRO
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
instrument for the use as primary outcome parameter in
clinical trials in UC which includes “clinically important signs
and symptoms of UC”. In clinical outcome assessment
research, a sign is an objective aspect of a condition or
disease that can be observed or measured.
A symptom is subjective from the patient point of view and
represents what the patient experiences about the condition
or disease; they are not able to be observed or measured
objectively. Signs can be evaluated using a PRO but are more
commonly evaluated/reported by the physician/investigator
who uses their medical training and judgement to score the
sign (a so-called Clinician-reported outcome, or ClinRO). The
Mayo score is an instrument that contains both signs and
symptoms, but the signs are evaluated and reported by the
physician/investigator (ClinRO), while the symptoms are
evaluated and reported by the patient (PRO).
152-159
Proposed change (if any): Can the agency confirm whether
the co-primary endpoint should be (a) symptomatic remission,
using PRO data on symptoms only, (b) remission of signs and
symptoms, as reported using a validated PRO which
encompasses both, or (c) remission of signs as reported using
a validated ClinRO and symptoms as reported using a
validated PRO?
Comment: It is encouraging that the EMA continues to
consider the measurement of patient-reported symptoms of
UC as a co-primary endpoint in establishing treatment benefit.
It is acknowledged that the patient perspective is key in
defining signs and symptoms which are important to evaluate
and concur with the Agency that a patient-reported outcome
(PRO) instrument to be used as primary outcome measure in
pivotal clinical trials in UC should be “completely and
rigorously validated”. When defining validity in PRO
instruments, the Reflection Paper on the regulatory guidance
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
for the use of Health-Related Quality of Life (HRQL) measures
in the evaluation of medicinal products
(CHPM/EWP/139391/04) alludes to validity, reliability,
responsiveness and interpretability for the specific
condition/setting. Numerous good practice guidelines have
been developed on PRO instrument validation since the
release of this guidance document (ISPOR, FDA etc) – can the
agency clarify the evidence they are looking for to support the
consideration of a PRO instrument as “completely and
rigorously validated”
152, 166-167
Proposed change (if any):
Comment: Please use “patient reported outcomes” and not
“patient related outcomes”
Proposed change (if any):
“Symptomatic relief should be evaluated by patient related
reported outcomes (PRO).”
152-159
“The use of this index may be justified, however, as previously
mentioned, an effect on both the patient related reported
sub-score and the endoscopic score is expected.”
Comment: The current text referring to symptomatic relief in
section 6.1 states...’ Symptomatic relief should be evaluated
by patient related outcomes (PRO). There are a number of
clinical indices, e.g. SCCAI (simple clinical colitis activity
index) mainly including patient reported symptoms. Whereas
these may be used provided that they are adequately
validated, this guideline recommends the further development
and validation of PRO instruments for the use as primary
outcome parameter in clinical trials in UC. Such an instrument
should include clinically important signs and symptoms of UC,
e.g. increased stool frequency and rectal discharge of blood.
An instrument to be used as primary outcome measure in
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
pivotal clinical trials in UC should be completely and rigorously
validated.’... This omits reference to the domains of the Mayo
Score that has been used previously in clinical trials, and has
ambiguity in its recommendation and should be updated.
164-167
164-168
Proposed Change: We proposed the following updated text.
‘Symptomatic relief should be evaluated by patient related
reported outcomes (PRO). There are a number of clinical
indices, e.g. SCCAI (simple clinical colitis activity index) and
Mayo Scoring Tool that include assessment of patient reported
symptoms but current signs and symptoms scales are not
considered adequately validated. This guideline recommends
the further development and validation of PRO instruments for
use as primary outcome parameter in clinical trials in UC.
Such an instrument should include clinically important signs
and symptoms of UC, e.g. increased stool frequency and
rectal discharge of blood. An instrument to be used as primary
outcome measure in pivotal clinical trials in UC should be
completely and rigorously validated.
Please note that the Mayo score is a composite scale. Is the
statement that co-primary endpoints (PRO plus endoscopy)
are required intended to remove use of a composite index that
includes both PRO plus endoscopy, such as the Mayo score?
Will sponsors now need to show impact on both parameters
separately? Is the use of the stool frequency and rectal
bleeding subscores as one co-primary endpoint and use of the
endoscopic subscore as another co-primary endpoint
recommended?
Comment: We advocate the continued use of the total Mayo
score (including PGA) until new PRO endpoints/criteria have
been validated. Furthermore, the continued collection of the
total Mayo score will be necessary to compare data collected
in active comparator studies (e.g. where the reference arm is
infliximab) where the historical data for the reference arm is
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
based on efficacy demonstrated using the criteria of the total
Mayo score.
169-170
172-192
Proposed change: Delete or preface the statement in lines
167-168 which discourages the use of the total Mayo score as
a primary interest in future studies with clarification of when
use of the total Mayo score might be appropriately
acknowledged.
Comment: Please indicate if the EMA considers endpoints that
include surrogate markers of inflammation suitable to use as
secondary or exploratory endpoints. Will the inclusion of
surrogate markers of inflammation lead to additional language
in the label?
Proposed change (if any): See above.
Comment: The revised guideline clearly indicates that
“remission” of signs and symptoms is the preferred way of
scoring the PRO used to support a co-primary endpoint. This is
contrary to a recent marketing authorisation approved by EMA
for Entyvio (vedolizumab) in May 2014 in which the primary
endpoint for induction of remission was clinical response
defined as a reduction in Mayo score of ≥ 3 points and ≥ 30%
from baseline and a decrease in the rectal bleeding subscore.
As the trial enrolled moderately to severely active UC patients
with a Mayo score 6 to 12, some patients would have been
defined as clinical responders based on less stringent criteria
than is being now proposed i.e. with an endpoint score >1.
We believe that this less stringent criteria, representing
“decreasing the severity of UC” employed in the Entyvio
development program remains clinically relevant for patients
and should be considered as an additional parameter on which
to define treatment success beyond remission.
Proposed change (if any): Agency to clarify if clinical
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
173
173-176
response can be accepted as primary endpoint
Comment: Achieving steroid free remission is an appropriate
endpoint in clinical trials and in clinical practice. However, it is
not an appropriate primary endpoint in induction, as (i) not all
patients will be on steroids on entry to the study, and (ii)
subjects will often remain on a fixed dose of steroids
throughout the induction period, with protocolized weaning
after achieving response or remission at the end of induction.
Thus, in many trial designs, it is impossible to achieve this
endpoint at the end of induction. The proposal ignores the
substantial clinical benefit that many subjects gain by
achieving clinical response. Furthermore, this proposal will
lead to unfeasibly large and expensive clinical trials and act as
a disincentive for conducting clinical trials in IBD, to the
detriment of this patient population with substantial unmet
need.
Proposed change (if any): Consider changing steroid free
remission to “an appropriate goal of treatment and an
appropriate secondary endpoint”.
Comment: Text in lines 173-176 does not include text
regarding induction/maintenance of a clinical response
‘Achieving/maintaining remission free of steroids is an
appropriate primary end-point. In patients receiving systemic
steroids these should be tapered according to predefined
schedules. For induction studies of short duration requiring
early evaluation of efficacy a low dose of steroids may be
acceptable provided that the dose is clearly justified and prespecified.’
Proposed change (if any):
Achieving/maintaining remission free of steroids is an
appropriate primary end-point. Alternatively
achieving/maintaining a clinical response based on a clearly
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
174-176
175
defined and agreed upon response criteria would be
considered as an appropriate primary endpoint. In patients
receiving systemic steroids these should be tapered according
to predefined schedules. For induction studies of short
duration requiring early evaluation of efficacy a low dose of
steroids may be acceptable provided that the dose is clearly
justified and pre-specified.
Comment: We agree, that when feasible, a low dose
corticosteroid is desirable for entry into clinical trials based on
several considerations including minimizing the treatment
effect due to the corticosteroids and reducing the potential
side effects of high dose steroids that are typically maintained
at baseline doses throughout the induction period. However,
we do not recommend exclusion of patients who require
higher doses of corticosteroids as this practice would have the
potential to exclude patients who have higher disease activity
and therefore limit the ability to understand the effectiveness
and safety of the therapy in this more severe population (Ha
et al, Clinical Gastroenterology and Hepatology, 2012,
12:1002-1007).
Proposed change: Delete reference to “low dose” and
restate as “….concomitant steroids would be acceptable
provided that the dose is clearly justified and pre-specified.”
Clarity is needed regarding the definition of a “short duration.”
It seems reasonable to allow steroids to remain stable during
induction studies of up to at least 12 weeks (at a dose of 20
mg/day or less) in order to avoid confounding due to
differential rates of taper in the treatment arms. It is not
necessarily appropriate to withdraw steroids in a patient who
has not yet demonstrated a response to study drug. Also,
strict tapering schedules are difficult to define and to
standardize given the multiple different corticosteroids
available.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
177-178
184-187
Comment: Please clarify that signs and symptoms refer to
stool frequency and rectal bleeding. Please add definition for
signs and symptoms, i.e. stool frequency (SF = 0-1) and
rectal bleeding (RB =0)
Proposed change (if any):
Signs and symptoms are defined by stool frequency (SF
= 0-1) and rectal bleeding (RB =0)
Comment: The guidance states :“Correspondingly, when
clinical symptoms are evaluated using the clinical part of the
Mayo score, a score of 0 or 1 may be used to define
symptomatic remission”
and
“Irrespective of scale used, the definition of remission should
encompass cessation of rectal bleeding”.
We believe that these two criteria might contradict each other.
A Mayo score of 1 could result from the bleeding score of 1 (=
“streaks of blood with stool less than half the time”),
indicating rectal bleeding.
188
Proposed change: Replace recommended definition of
remission with the following: “Clinical remission could be
defined as a Mayo score ≤ 2 points, with bleeding score=0
and no other individual subscore > 1”.
Comment: The Mayo Clinic Score is composite within a
patient.
Please clarify whether symptomatic remission and MH is a coprimary endpoint (at population level) or a composite endpoint
(within the same patient) and the timing of assessment.
Please also clarify whether re-randomization into maintenance
is based on a patient achieving either or both symptomatic
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
remission and/or MH.
Please also clarify if co-primary endpoints may be assessed at
different time points.
188
Proposed change (if any):
“As outlined above, symptomatic remission and MH should be
considered composite co-primary endpoints.”
The Agency’s Draft UC Guidance suggests a “Co-Primary”
endpoint approach that considers both symptomatic relief as
well as an effect on the inflammatory process rather than the
more commonly used Composite indices, such as the Mayo
Clinical index.
“A significant effect on both aspects of the disease is required
(co-primary endpoints). Composite indices including both
symptoms and MH, such as the Mayo Clinic index have been
used in several clinical trials. The use of this index may be
justified, however, as previously mentioned; an effect on both
the patient related sub-score and the endoscopic score is
expected. It has to be stressed that the total Mayo score
including physician’s global assessment is not of primary
interest.”
“As outlined above, symptomatic remission and MH should be
considered co-primary endpoints. However, as listed below,
achieving both symptomatic remission and MH (for the
individual patient) is considered an important secondary
endpoint. “
In the FDA’s Ulcerative Colitis: Clinical Trial Endpoints
Guidance for Industry released Aug 2016, a composite
endpoint which includes an assessment of signs and
symptoms and endoscopic improvement.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
“We currently recommend a primary endpoint of clinical
remission (responder definition based on Stool Frequency,
Rectal Bleeding, and Endoscopy scores) (see section IV.,
Interim Approaches to Efficacy Assessments). Until a valid
patient-reported outcome instrument for UC signs and
symptoms and a valid clinician rating scale for mucosal
inflammation in UC become available, a modified Mayo or
modified UCDAI score omitting the physician’s global or
disease activity ratings, as described in section IV, can be
used as an endpoint measure.” (FDA UC Guidance, Sect V.A.)
195
195
It is Lilly’s opinion that the diverging definitions of the primary
endpoints could prove to be problematic and a shift to more
consistent desired primary endpoint definition between
regional guidance would be of benefit to patients, payers, and
sponsors.
Comment: Clinical response (or endoscopic response) is used
as a primary endpoint in most published (and ongoing) clinical
trials. These are clinically relevant endpoints and the numbers
required to show these endpoints are feasible for phase 2 and
registration programs.
Proposed change (if any): Please consider amending the
guidance to recommend the use of clinical response or
endoscopic response as a suitable primary endpoint in UC
clinical trials.
Comment: Please consider specifying the change in the
individual components of a disease activity index that would
be appropriate to meet a definition of clinical response or
clinical remission. This is included in the FDA draft guidance
and was discussed at the FDA’s clinical endpoints conference
and the GREAT3 conference.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
197-200
201
201-201
201-204
Proposed change (if any): Please specify examples of the
changes in disease activity index scores required to achieve
clinical response or clinical remission, to provide consistency
with FDA guidance.
Comment: The revised guideline clearly indicates that
“remission” of signs and symptoms is the preferred way of
scoring the PRO used to support a co-primary endpoint. This
method provides evidence at an individual level (i.e.
proportion of responders). However, group level changes (e.g.
mean change from baseline) in the PRO will be additional
useful information to understand the overall efficacy of an
experimental treatment, providing additional data through use
of a continuous variable beyond that which can be understood
through the creation of a binary outcome. As such, we would
like to encourage the Agency to include group-level responses
as key supportive/secondary endpoints, (unless already
covered under the line 200 “numerical evaluation of the
symptoms score”=
Proposed change (if any): Please indicate whether the
endpoints can be analysed as change from baseline
Please clarify what scales are appropriate for histological
evaluation and at what time point these evaluations are
expected
Comment: We consider that histological improvement would
be an additional valuable secondary endpoint.
Comment: We acknowledge that biopsies for histologic
disease activity would be collected at screening and posttreatment to assess for subsequent histological healing.
However, there is no standardized histologic scoring system,
nor is there one that has been validated (Peyrin-Biroulet, L et
al. American Journal of Gastroenterology (2015)110: 13241338).
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
203
203-204
Proposed change (if any): Evaluation of histological
improvement should be included as an exploratory endpoint to
assess UC activity and treatment efficacy.
Comment: Elevated C reactive protein (CRP) is not a common
feature of UC, in contrast to Crohn’s disease. This is not a
tractable secondary endpoint.
Proposed change (if any): Consider removing.
Comment: Please clarify if reference is made to faecal
calprotectin and not serum calprotectin? If serological, we
recommend this being an exploratory endpoint and not
included within the composite endpoint, as there is limited
data on this.
Please clarify what is meant by normalisation of CRP, as many
patients do not have elevated CRP levels.
Please clarify what label claim this endpoint supports.
208
Proposed change (if any): “Patients achieving MH, judged
endoscopically, as well as combined clinical,
other biomarkers for inflammation serological (eg,
=normalisation of normal CRP and/or faecal calprotectin)
and histological remission”.
Comment: The revised guidance proposes the use of a
validated quality of life (QoL) measurement to support a
secondary endpoint related to changes in QoL. The Agency
gives examples of both IBD-specific and generic PRO
instruments to support this endpoint – does the agency have a
preference for one over the other? Most endpoints listed in
section 6.1.1.2 provide information on preferred
analysis/presentation of results; the QoL endpoint does not.
Unlike the Mayo score in which remission can be clearly
defined by the absence of signs and symptoms, none of the
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
instruments included in the example have clinically defined
criteria for optimal outcomes. As such, can the agency advise
on how the endpoint should be developed i.e. mean change
from baseline, proportion of patients achieving a pre-defined
clinical benefit etc?
212
213-215
213-215
Proposed change (if any): Please indicate any preference
the agency may have in terms of endpoint and preferred
analysis (e.g. SF-36 and EQ-5D)
Comment: Colectomies are infrequent in 12-52 week clinical
trials, particularly when subjects are allowed rescue therapy.
Consequently, reduction in the number of colectomies may not
be a suitable secondary endpoint for UC clinical trials, outside
of the proposed indication of “treatment of acute, severe UC”.
Proposed change (if any): Consider removing.
Comment: The suggested endpoint of corticosteroid-free
remission may be challenging to meet in this difficult-to-treat
sub-group and require an unfeasibly large study to have
sufficient power to meet this endpoint. Clinical response
would be a more feasible primary endpoint in these patients,
with corticosteroid-free clinical response as the first secondary
endpoint.
Clinicians treating patients with UC understand the difference
between these endpoints and the implication of achieving each
of these endpoints on a per-patient basis.
Proposed change (if any): Consider changing corticosteroid
free remission from a primary to a secondary endpoint for this
cohort of patients. Consider avoiding a recommendation for a
sub-group analysis of corticosteroid-free remission
Comment: The minimum duration of clinical remission in the
absence of steroids required to achieve an endpoint of
corticosteroid-free remission is not specified. Feedback from
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
key opinion leaders suggests that a minimum period of 4
weeks is both clinically relevant and important.
218
273
275-276
284
Proposed change (if any): Consider making a statement on
the minimum duration of steroid-free remission (or response)
required to meet these endpoints.
Comment: Please clarify the definitions for “mild, moderate
and severe”, and how these are measured (i.e. anatomical
location versus the Mayo score versus a composite of both).
We consider the Mayo score is more appropriate than the
anatomical location (i.e. mild=1, moderate=2, severe=3).
Proposed change (if any):
“…e.g. mild, moderate and severe by MCS, ES, or
UCEIS.”
Comment: Stipulates that subjects entering trials must have
had “recent visualization of their GI tract” but does not
stipulate what is meant by “recent”.
Proposed change: Propose recommendation for what is
considered "recent".
If the Full Mayo score is no longer accepted (as noted on line
167-8, which states the PGA component is no longer of
interest), then please update this sentence to indicate what is
acceptable to define disease for inclusion. Also, is the range
specified in this sentence intended to define only moderate to
severe disease?
Comment: We do not agree with the recommendation that
the primary endpoint of an induction study should be steroid
free remission.
The reasons to maintain stable corticosteroid doses during the
induction period include the following:
There would be insufficient time to taper corticosteroids prior
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
to the primary endpoint assessment using the type of tapering
schedule generally applied in UC clinical trials.
A rapid corticosteroid taper prior to the primary endpoint
assessment may precipitate clinical flares that would impact
patient well-being and could present challenges to the
interpretation of the treatment effect during the induction
period. Specifically, a rapid taper of corticosteroids during
the induction period could confound the assessment of efficacy
in the setting of additional medication changes.
Furthermore, a rapid steroid taper may introduce an
imbalance in efficacy in the Placebo vs active treatment group
that could result in lower efficacy in the PBO group and would
confound assessment of efficacy.
Withdrawal of corticosteroids prior to the induction primary
endpoint could also lower the number of patients that may be
ultimately eligible for the maintenance study.
In some UC clinical trials, corticosteroid tapering is mandatory
in clinical responders using defined criteria over a longer time
period during the maintenance period.
Subgroup analyses of induction and maintenance UC trials
demonstrated that patient steroid status at study entry did not
influence the ability to achieve response or maintain response.
These results support the conclusion that meaningful
information can be obtained with steroid tapering initiated
during maintenance treatment to demonstrate the benefit of
the active study treatment vs. Placebo for achieving and
maintaining clinical remission.
Proposed change: Delete reference to steroid taper during
induction.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
Additional
related
comments:
262-263
Mandatory steroid tapers during induction periods of short
duration introduce the potential for provoking severe flares in
patient who are already ill. Please allow sponsors the ability
to keep steroids stable during induction studies. It is
understood that labelling would reflect the study design. This
sentence adds to the confusion about whether steroid-free
remission is an expected primary endpoint or there is the
option for it to be a secondary endpoint.
284
Here the document again suggests that the primary endpoint
should be steroid-free remission, which is inconsistent with
some earlier statements in the document and it is not clear
which co-primary endoscopic endpoint is acceptable.
In active comparator studies with a placebo, please clarify
which comparison is expected to support approval—versus
placebo? Or the active comparator?
Formal comparisons to active comparators are logistically not
very feasible in a pivotal trial exploring various dose regimens
as well as a placebo for proper safety evaluation; thus should
not be a requirement for regulatory approval; informal
“reference” arms with an active comparator are prone to alfa
and beta errors and may rather cause confusion, unless strong
superiority is expected for the IMP and should not be
requested.
Add-on of IMPs to TNF inhibitors may be difficult to justify
ethically due to safety concerns and would not allow proper
safety evaluation for compounds with improved safety profile
over TNF antagonists
Comment: In maintenance of remission trials, recommend
that patient who are presently on the test drug should be rerandomized to continue the test drug or switch to placebo.
286-292
318-319
324-326
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
335
335-339
These are patients who were failing their standard of care
drugs and thus entered the trial; were induced, went into
remission, and now are entering the maintenance phase. Is it
ethical to re-randomize these patients to placebo in a waxingwaning chronic disease that is never cured without colectomy
and that is known to recur off medication?
Comment: The guidance states that “Patients who are
steroid free remission (as defined above) are eligible for
inclusion into the trials". As indicated in the response to line
284 of the guidance above, we have concern regarding the
tapering of steroids during a 6-8 week induction study &
therefore, this concern carries over to the definition of the
target population for maintenance studies. We advocate for
the target population of a maintenance study to include
patients who achieve a pre-specified measure of clinical
response as this represents the broadest population of
patients to be treated in the clinical setting. Among these
patients will be those achieving clinical remission both on and
off steroids who can then be the target populations for major
secondary analyses for maintenance of clinical remission and
steroid-free remission with appropriate statistical controls.
Proposed change: Acknowledge that patients in clinical
response are an appropriate primary target population for the
assessment of maintenance therapy.
Requirement for inclusion into maintenance trials of only
patients who are steroid-free is problematic, as 8-12 week
induction trials may not be sufficient in duration to wean all
subjects completely off steroids. Furthermore, continuation of
treatment in patients only who have reached a stringent
endpoint of remission, steroid discontinuation AND mucosal
healing is not consistent with the clinical paradigm of treating
patients with at least a partial PRO and/or endoscopic
response to induction treatment. This requirement will also
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
335, 340
have a large impact on the side of induction trials needed in
order to identify adequate numbers of subjects for
maintenance trials.
Comment: Only patients who are in steroid-free remission
are eligible for inclusion into the maintenance phase. This is
not a technically achievable outcome at the end of induction,
as patients who enter the induction period on corticosteroids
will most likely remain on steroids for the duration of this
period and have protocolized weaning when they enter
maintenance.
Only patients in remission should enter maintenance. This is
highly restrictive and would result in unfeasibly large induction
studies. Including patients with a clinical response into
maintenance studies reflects real-world clinical practice and
allows sponsors to address important, clinically relevant
questions, such as determining the proportion of responders
who enter remission.
Sponsors should also be encouraged to allow non-responders
to continue on active medication in limited circumstances
where the sponsor predicts that subjects are likely to have a
“late response” to the study drug, e.g. Ustekinumab’s UNITI
phase 3 studies in Crohn’s disease.
A proposal is to allow patients with symptomatic response to
be included in the maintenance study with a stratification
factor of remitter and responders. The primary endpoint of
this maintenance study would be based on both responders
and remitters. Subgroup analyses would be provided to show
consistency between subgroups.
Proposed change (if any): Consider changing these lines to
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
340-353
reflect that patients in clinical response should be allowed to
enter maintenance studies.
Comment: Also refer to comments in response to Lines 136141.
The notion that true maintenance of efficacy can only be
demonstrated in the context of a randomized-withdrawal
study (vs. placebo) or only among induction
responders/remitters is concerning. As discussed in an earlier
section, the arbitrary designation of induction and
maintenance study periods limits one’s ability to evaluate the
true efficacy potential of a MOA; and is highly inconsistent vs.
clinical practice.
The maintenance of efficacy among “induction responders”
only provides insights into the continued benefit observed
among patients who achieved an initial response/remission
within an arbitrarily set “early” timeframe, but ignores the rest
of the population treated. Whereas, the holistic approach
under a treat-through study design, will support the
evaluation of long-term efficacy at a population level,
including both early and late responders to initial (induction)
treatment and their response to continued long-term
treatment (maintenance), and will also support the desired
“maintenance of remission among induction remitters”
analysis.
In addition, evaluation of endoscopic/ histologic endpoints
would be significantly challenged in the setting of a
randomized-withdrawal (to placebo) study, since the kinetics
of disease worsening (upon discontinuation of treatment) by
these outcomes measures are unknown. A treat-through study
design is much more favourable and preferred for the
evaluation of these important outcomes.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
It should be noted that comparison to standard of care
comparators (e.g. anti-TNF) using this methodology incurs
substantial complexity. We believe comparison to SOC in both
induction and in maintenance phases of treatment as part of
the confirmation study is best accomplished using a treat
through methodology.
355-356
Finally, the validity or requirement of a randomized withdrawal
(to placebo) design to demonstrate the need for maintenance
treatment in patients with UC should be questioned. After 20
years and numerous trials across different MOAs, there is no
evidence that patients with UC can be successfully managed
without active maintenance treatment. All of the randomized
withdrawal studies of biologic agents have demonstrated the
need for continued maintenance treatment. It should also be
noted that randomized withdrawal placebo studies are
inconsistent with clinical practice and is a design feature that
is a significant deterrent to patient recruitment.
Comment: The guidance recommends that the primary
endpoint of maintenance studies should be steroid-free
remission maintained without surgery. In clinical trials
colectomy is considered one of potentially several possible
treatment failures in the analysis of efficacy; therefore study
patients who achieve or maintain remission at the end of
study have not undergone colectomy. Further, pursuant to
comments related to line 335 above, we believe that clinical
remission among subjects induced into clinical response
represents the broadest evaluation of maintenance therapy
and should be the primary endpoint of a maintenance study.
Pre-specified major secondary endpoints of maintenance of
clinical remission and steroid-free remission based on
appropriate subgroups would provide additional important
measures of the effectiveness of a maintenance therapy.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
Furthermore there has been a generally low incidence in
some UC populations, including an unselected UC population
in Europe (Hoie O, eta l (2007) Gastroenterology 132:507515) as well as what was observed in the ACT studies of
infliximab (Sandborn WJ (2009) Gastroenterology. 137:12501260) that the incidence of surgery is relatively low of a 1
year period. Therefore, it would be a challenge to power a
clinical study for a steroid free remission endpoint that
includes the absence of surgery.
355
370-372
371-372
Proposed change: Update recommendation on the primary
endpoint to clinical remission among subjects responding to
induction treatment with major secondaries focused on the
subgroups of subjects who maintain clinical remission or
achieve steroid-free remission during maintenance therapy.
Different primary EP then recommended before (lines 175,
284); also the 1yr-surgery rate is too low to see a significant
reduction in a typical trial population and setting
Please provide guidance how a maintenance trial is to be done
for comparators that are indicated only to be continued in
induction responders (i.e., anti-TNF agents). Parallel group
designs are problematic in that, per label, patients without
response to the anti-TNF agent (or integrin inhibitor) are not
to continue the drug. This could lead to differential drop outs
between arms that, as the document has already indicated,
are methodologically problematic. On the other hand, it
seems inappropriate to take patients who responded to one
drug and randomize them to a different agent for the
purposes of testing “maintenance.” Much more guidance is
needed here, as sponsors are really struggling with these
questions.
Unclear how an AC should be incorporated in a randomized
withdrawal design; that would require to switch e.g. TNF-
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
382-392
395-396
431 – 433
431-437
refractory pts responding to induction with the experimental
drug to TNF or a different drug of unknown activity for that
patient. Ethically very difficult and not acceptable to many
investigators and patients.
The discussion of refractoriness in to other drugs in a
maintenance setting is unclear. Is this section intended to
refer to the concomitant/prior meds prior to induction with
whatever agent was used prior to putting the subjects into a
maintenance trial?
Comment: Please clarify if the one year maintenance phase
can include 12 weeks of steroid tapering at the beginning of
the maintenance phase.
We consider 6 months as appropriate as one year for steroidfree remission, and therefore recommend that patients must
be steroid free for 6 months prior to the one year primary
endpoint assessment. We propose to allow a single, shortterm steroid dose due to other, unrelated conditions.
Comment: As the draft guideline states, UC is rare below 10
years of age. Yet the clinical development is asked to include
patients from 2 years of age. Fully-powered clinical efficacy
studies in the paediatric population 2 – 10 years of age might
not be feasible due to the low patient numbers. Therefore
clarification is needed on what is expected to be demonstrated
in those children. (see also next comment)
Comment: The comment that children from 2 years of age
and older should be included in clinical development programs
requires clarification. The key point here is the age at which
the subject’s IBD was diagnosed, which is inversely
proportional to the likelihood that the subject has a rare,
monogenic cause for IBD. We would advocate that the age at
diagnosis of patients that should be included in pediatric IBD
clinical trials is 7 and above, which is consistent with the
current definition of “Very Early Onset IBD” (VEOIBD, patients
with IBD onset <6 years of age). Furthermore, even though
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
the draft guidance discusses testing for monogenic defects
that may cause IBD, it states that subjects can be included or
excluded based on the defect. This guidance is confusing, as
it appears to be mandating the inclusion of pediatric subjects
with rare, monogenic causes for IBD, in pediatric clinical trials
that are designed to investigate idiopathic IBD.
436-437
443-463
Proposed change (if any):
1) Consider rewording this section to base the pediatric
subject’s age on the age at diagnosis, rather than the current
age of the subject.
2) Consider explicitly using the term “Very Early Onset IBD” to
make it clear that the intention of this guidance is not to
mandate the inclusion of pediatric subjects with rare,
monogenic causes of IBD in trials designed to include subjects
with idiopathic IBD. The Agency should consider
communicating a clear expectation that rare, monogenic
causes of IBD will be considered orphan diseases.
Comment: Please clarify the term “younger children” by
adding an age.
We agree to genetically testing children, but should not be the
sponsor´s burden.
Proposed change (if any):
“Younger children <6 years of age should be should have
been genetically tested for known immunological defects
and in-or excluded depending on the defect.”
In designing clinical trials, extrapolation (particularly for
conclusions of efficacy) implies that the adult trial leads the
paediatric trial. Only then can extrapolation result in a
reduction in the amount of data required. While the goal has
been to have a concurrent adult/paediatric development, this
benefits a staggered development more than the latter. If a
concurrent development is pursued (at least with
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
adolescents), the concept of extrapolation, even if it is
conceivable (say, same pharmacological class), seems to be
not helpful.
488-489, 492-493
490-491
Proposed change (if any): the guidance should explain how
extrapolation, where conceivably applicable, can be used in
concurrent development; endpoints for adults and paediatrics
are not the same e.g. PUCUI; time points of assessments are
not the same
Repeated endoscopies in children are problematic. Children in
particular have more issues with the preparatory regimens
than adults do, as these and the procedure itself interfere with
school and activities. Please consider allowing a symptom
related endpoint (PUCAI) at least for induction endpoints (i.e.,
select either induction or maintenance where endoscopy is
required but not both). Paediatric UC, as an orphan disease,
is already difficult to enrol into clinical studies due to the rarity
of the disease. Multiple invasive procedures deter enrolment
into clinical studies.
Comment: Please clarify, if clinical response/remission alone
could be an acceptable primary endpoint for induction, if
combined with endoscopic MH for maintenance? This would
allow for having fewer endoscopies in children.
For trials with both an induction and maintenance phase, is it
acceptable to perform endoscopy at the end of the
maintenance phase and not at the end of induction?
Would it be acceptable to separate endoscopy from the
induction/maintenance paradigm (i.e. perform endoscopy at
the 6 month time point)?
Or Please clarify, if endoscopy has been shown in adults, can
it be waived in children?
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
504-519
505
515-516
520-521
Please clarify the circumstances when it is acceptable to waive
endoscopy within a trial? Can endoscopy be waived in certain
age groups (i.e. under 12 years of age)?
When extrapolation is not possible, a non-inferiority trial
cannot be operationalized in terms of the choice of the noninferiority margin.
Proposed change (if any): explain whether the margin can
be based on the adult trials (from which extrapolation cannot
be used) or other paediatric trial in the same indication.
Comment: Please consider the current joint
ESPGHAN/ECCO/PIBDnet/Canadian Children IBD Network
position statement on placebos in pediatric IBD clinical trials.
Turner et al. J Ped Gastroenterol Nutr 2016. Pediatricians will
only support the use of placebos in pediatric IBD studies
where there is genuine equipoise between active treatment
and placebo.
Proposed change (if any): The paragraph on placebos in
pediatric studies may have to be re-written to reflect current
expert guidance.
Comment: Please clarify, what is the risk of “lack of efficacy”?
Proposed change (if any):
“In case the use of placebo control group is considered
necessary, where there is no data from adults, all efforts
need to be made to assure that the patient is not exposed to
more than minimal risk”.
Comment: The guidance states that in pediatric trials,
combined induction and maintenance trials can be accepted
(as opposed to what was stated for trials in adults). Are these
combined trials in children allowed without re-randomization
to continued study drug vs. withdrawal to placebo in the
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
536-538
542-543
Sections 8.3.1.1 –
8.3.1.5
Please add more rows if needed.
maintenance phase? Please provide further guidance.
Comment: Not all mechanisms of action for the treatment of
Ulcerative Colitis may impact adaptive immunity. If preclinical
data exist demonstrating that vaccination responses are not
affected this should suffice.
We suggest removing the requirement that studies evaluate
impact on vaccination of all drugs with new mechanism of
action, and limit to drugs interfering with adaptive immune
response only where preclinical data suggest increased risk of
failed vaccination.
Comment: Please clarify “ if a cross company registry” or a
“cross paediatric GI registry established by a professional
organisation such as ECCO” is intended
Comment: Further clarification on whether these sections
apply to above 10 years old patients’ needs to be provided.
Older adolescents (14 and older) could be included in the
adult development studies. If the sections apply to 2 – 10
year olds (as understood per introductory statements to
Section 8.3.1), such studies may face feasibility issues.
3. Specific comments on text – Minor Comments
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
11
32
107
182-183
201-202 vs 204
211
260-262
Comment: Suggest to add ‘ulcerative colitis’
There is in reference to “Pharmacodynamics” as section 7.1.2
in the Table of Contents but there is no text/information in the
document regarding this topic in section 7, where 7.1.2
describes the topic “Interactions”
Comment: “includes pancolitis”. E3 distribution, involving the
colon proximal to the splenic flexure is pancolitis, by
definition. See Satsangi et al. Gut 2006 55(6): 749-53.
Proposed change (if any): Delete “includes”
Please describe how “standardization of reading should be
convincingly demonstrated” can be demonstrated (what is
expected?)
Comment: Please define “histological normalisation”.
What is the difference between “histological normalisation”
and “histological remission”?
We suggest to use the same terminology.
Comment: Please clarify how steroid sparing effect is different
from the primary endpoint.
Comment: Appropriate follow-up period off therapy is
recommended to see if patients who are in remission at the
end of treatment remain in remission at the end of follow-up.
However, with the exception of corticosteroids, there is no
medication for UC that is withdrawn (unless there is an
adverse event) even when the patient is in remission. This is a
chronic waxing and waning disease, and there is ample
evidence that withdrawal of maintenance medication results in
an increased risk of relapse and greater difficulty in reinducing remission. In addition, with certain biologics,
withdrawal of therapy and restarting may increase the risk of
developing antibodies to the drug and reduce its effectiveness.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
262-264
280-281
325
340-346
Comment: “Patients on steroids at entry should have their
dose tapered according to predefined tapering schedules.
Obtaining steroid-free remission should be the goal of
therapy.”
Proposed change (if any): Consider carefully distinguishing
between corticosteroid-free remission as a well-established
goal of therapy in clinical practice and corticosteroid-free
remission as an endpoint in clinical trials. The above
sentences may lead to confusion in this regard.
As patients who are newly diagnosed may be appropriate for
inclusion in studies (i.e. Bionaive patients), we do not believe
a 3 month period from diagnosis is appropriate. We suggest
considering a provision of 3 months of symptoms prior to
diagnosis in newly identified patients.
Comment: “Patients who are presently on the test drug
should be randomised to continuing the test drug or switching
to …”
It seems clear from the chapter 7.2.2.2.2 below that rerandomisation is what is meant in this sentence
Proposed change (if any): Patients who are presently on
the test drug should be re-randomised to continuing the test
drug or switching to …
Comment: text on lines 340 -346
Trials combining induction treatment and maintenance
treatment should preferably only enter patients that have
achieved remission (in either the trial drug or comparator
group), into the maintenance phase. Responders may be
included in the maintenance phase as it is considered relevant
to study if continued treatment in responders may eventually
lead to remission. However, if the intended claim is
“maintenance of remission”, the primary analysis should be
based on the remitters only.
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
393-394
400-401
402
402-404
Proposed change (if any):
Trials combining induction treatment and maintenance
treatment should preferably only enter patients that have
achieved remission (in either the trial drug or comparator
group), into the maintenance phase. Responders may be
included in the maintenance phase as it is considered relevant
to study if continued treatment in responders may eventually
lead to remission or if maintenance of response is an intended
claim. However, if the intended claim is “maintenance of
remission”, the primary analysis should be based on the
remitters only.
The tapering schedule given should specify the pertinent
group of steroids. As commented earlier, there are different
groups/compounds which may require different schedules.
Comment: For the reasons that the author alludes to on lines
398-400, topical rectally administered therapies are usually
excluded from industry-sponsored clinical trials in UC. Given
this fact, the statement on lines 400-401 is confusing.
Proposed change (if any): The sentence on lines 400-401
should be amended or removed. Clarification is required.
Comment: Treatment with antibiotics should be at the
discretion of the sponsor.
Proposed change (if any):
“Antibiotics should normally be excluded and at the
discretion of the sponsor and in severe disease, anticholinergic, anti-diarrhoeal, NSAID and opioid drugs should
not be allowed as they may contribute to worsening of the
relapse”.
It is infeasible to exclude antidiarrheal and pain medications in
a 52 week study in this patient population. Exclusion of these
medications will greatly limit enrolment and limit
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
422
431
488
488-489
492-493
495-496
generalizability of studies in patients with UC
Comment: “Furthermore, it is important to get information on
re-treatment outcomes even after a longer time interval
without treatment with a specific drug.”
Given this information may take long time to collect, it should
not be a requirement for the initial submission. Please specify
whether this information can be provided post marketing
Proposed change (if any): Furthermore, it is important to
get information on re-treatment outcomes even after a longer
time interval without treatment with a specific drug, this
should be considered as part of post marketing
commitment
Comment: Please clarify that paediatric Ulcerative Colitis is a
rare disease in younger children.
Proposed change (if any):
“Paediatric UC is a rare disease and younger in children
(i.e. under 4) below 10 years of age may develop a different
disease phenotype compared with adolescents or adults.”
Comment: Please clarify that clinical remission is the same
for both adults and children.
Comment: Please clarify that clinical remission and
endoscopic MH could be separated in time.
How do you re-randomize, based on the co-primary endpoint?
Can you re-randomize based on response?
Comment: Please clarify that endoscopy can be performed
within a subset of patients.
Comment: In a paediatric UC population, the revised
guideline proposes use of the PUCAI as a surrogate for
symptomatic remission. However, there are two versions of
the PUCAI – one which is self-reported by the patients (PRO)
and another which is reported by the physician/investigator
(ClinRO). Guidelines from ISPOR (Matza et al., Value in Health
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
16 (2013) 461 – 479) suggest that reliability of responses to a
PRO cannot be assumed before the age of 8 (assuming no
cognitive functioning deficits). Given that the agency proposes
that “the clinical development program should include children
from 2 years of age and older”, does the agency have any
advice on when to utilise the PRO and the ClinRO version of
the PUCAI?
498-499, 500-503
510-514
530-532
Proposed change (if any): clarify which version of the
PUCAI EMA recommends for paediatric studies
“Sustained relapse-free steroid-free remission” is a stringent
endpoint (more stringent even than the endpoint advocated
for adults” and may discount benefit in subjects who achieve
remission but are not entirely able to discontinue steroids or
who suffer a brief relapse but subsequently improve. It is
acknowledged that steroid-free remission is an appropriate
endpoint for a given patient, it is potentially overly stringent
for a clinical trial endpoint, especially in a heavily treatment
refractory population with a large unmet need.
The acknowledgement regarding the limitations of placebo use
in paediatric UC patients is appreciated. However, the
suggestion that a NI study against an active comparator are
reasonable replacements runs counter to the concept of
extrapolation (whereby similar of effect in paediatric subjects
needs to be demonstrated in a drug that has already proven
efficacy and safety in adults) and are infeasible due to the
need for relatively large samples sizes (in an orphan disease)
and the potential need for a double-dummy design (in the
case of differing routes of administration between study
drugs).
Comment: Please specify “development”. We propose to
change the wording to “growth velocity”
Proposed change (if any):
Line number(s) of
the relevant text
(e.g. Lines 20-
Stakeholder number
Comment and rationale; proposed changes
Outcome
(To be completed by
(If changes to the wording are suggested, they should
(To be completed by the Agency)
the Agency)
be highlighted using 'track changes')
23)
530-532
543
“Post-study/post authorisation long-term data, either while
patients are on chronic therapy or during the post-therapy
period, are necessary to determine possible effects on
maturation and development growth velocity”.
Does this need for long-term safety data in growing children
mandate the need for post-approval registries?
If registries are established and are disease (and not drug)
based, does each company with a new drug have to establish
such a disease-based registry or can they collaborate and
form a single disease-based registry with patients on multiple
drug regimens?